Ran the Webasto all night in -3°C — battery bank didn't cope as well as I expected

Took the motorhome out to the Cairngorms last weekend. Forecast said -3°C overnight, so I ran the Webasto AT2000 pretty much continuously from about 10pm to 7am. Figured my 200Ah LiFePO4 (two Fogstar Drift 100Ah in parallel) would handle it no bother.

Woke up to the Victron BMV-712 showing 18% SOC. The Webasto pulls around 10W on low flame but spikes to 25W on startup and during the glow phase — I'd accounted for that. What I hadn't properly accounted for was the diesel pump cycling more frequently than usual in the cold, and the van's 12V fridge not switching off overnight like I'd assumed it would.

Ran the numbers the next morning. The heater alone probably drew around 120–130Wh across nine hours, which is fine. The fridge running continuously added another 60–70Wh on top. Still shouldn't have dropped that low — I'm wondering if the cold itself is clipping the usable capacity of the cells even though LiFePO4 is supposed to handle low temps better than AGM.

Has anyone else noticed actual capacity loss from their LiFePO4 at sub-zero temps, even when the cells aren't being charged? Discharge performance specifically. I've seen the spec sheets but spec sheets and a frozen Scottish hillside are two different things.

@RustyTinker the Webasto itself isn't the killer — it's the control unit and glow pin cycling that hammers the amps at startup. If it's kicking on/off repeatedly through the night trying to maintain temp, those startup surges add up massively.

Also worth checking your BMS low-temp cutoff. Fogstar cells don't like charging below around 5°C — if your alternator or solar tried topping up whilst the bank was cold, the BMS may have been throttling or disconnecting, leaving you with less usable capacity than you thought.

I run a static caravan setup rather than a van, but same principle applies — my Victron SmartShunt showed me some eye-opening figures the first winter I monitored properly. The actual consumption vs expected consumption gap was sobering.

What's your insulation like on the battery compartment? That alone makes a significant difference at -3°C.

@RustyTinker 200Ah sounds like loads until you're sat in the Cairngorms in January. The Webasto fuel pump ticks along fine but your parasitic loads are quietly murdering you — fridge still running, lighting, phone charging, whatever else you've left on.

Also worth noting: LiFePO4 BMS will start throttling or cutting output once cell temps drop. If those Fogstar batteries aren't insulated and sitting in an exposed locker, they'll have been cold-soaked all day before you even started the night. Cold cells = reduced usable capacity, simple as.

I run 300Ah in my setup and I'd still be cautious doing a full Scottish winter night without shore power or a DC-DC from the vehicle alternator as backup.

Check your actual consumption with a Victron BMV or similar — bet you're pulling more than you think.

Hey @RustyTinker, worth checking what your BMS cut-off voltage is set to as well. LiFePO4 cells don't love being discharged hard in very cold temps — most BMS units will throttle or trip earlier than you'd expect once the cell temperature drops. Your 200Ah nominal capacity effectively shrinks in those conditions. Did you have any battery heating pads fitted? The Fogstar Drift cells are reasonably cold-tolerant but they're not magic. Also worth looking at whether you had anything else quietly drawing power overnight — hab lights, a 12v charger left on, fridge compressor cycling. Those background loads stack up over nine hours more than people realise. Might be worth fitting a proper battery monitor like a Victron BMV if you haven't already, so you can see exactly what's happening rather than guessing afterwards.

@RustyTinker I've had similar on the narrowboat in a cold snap. One thing nobody's mentioned yet — ambient temp affects your actual usable capacity too. LiFePO4 doesn't drop off as badly as lead-acid in the cold, but it's not immune. If those Fogstar cells got properly chilled overnight you'd have lost some capacity on top of the draw.

Worth insulating around the battery bank if you haven't already. Even just some closed-cell foam sheeting made a noticeable difference on mine.

@RustyTinker one thing nobody's mentioned yet — at -3°C your alternator charging on the drive up there would've been less effective too, so you may have arrived with less in the bank than you realised. Also worth looking at your heater's ventilation setting if it has one; the AT2000 has a reduced power mode that cuts consumption noticeably once the van's up to temperature. I'd also double-check your cable runs to the battery bank — any voltage drop under sustained load means the BMS sees a lower voltage than you'd expect and can trip out prematurely. Two Fogstar 100Ah units should theoretically handle a Cairngorms night, but only if everything upstream is sorted. What's your cable gauge and run length out of interest?

Really interesting thread this. @RustyTinker one thing worth adding — the AT2000 draws quite variable current depending on what mode it's running in. At full blast it can pull around 10A during the glow plug warm-up phase, then settles lower, but if it's cycling on and off repeatedly through the night trying to maintain temperature at -3°C, those startup spikes add up considerably over nine hours. Worth logging your actual consumption with a shunt monitor if you haven't already — you might be surprised how much those repeated ignition cycles are costing you compared to a steady run. Also, have you checked your cable sizing between the batteries and the heater? Undersized cables cause voltage drop which can trigger the BMS earlier than you'd expect, especially when the AT2000 demands that initial surge current.

@RustyTinker had almost identical situation in the motorhome last February, parked up in Perthshire. The bit that caught me out was the Webasto cycling more frequently than expected because the hab space was losing heat faster through the floor than the walls — worth checking if you've got any insulation gaps underneath.

Also worth logging actual draw with a shunt if you haven't already. My Victron BMV-712 showed the AT2000 peaking noticeably higher during ignition cycles than the quoted average figures suggest. Those spikes add up over 9 hours.

200Ah usable on LiFePO4 is still only ~180Ah in practice once you factor BMS headroom.

Good points all round here. One thing I'd add — have you checked what your Webasto's actually doing at startup? The AT2000 has a pretty hefty glow plug draw for the first couple of minutes, something like 10-15A, and if that's cycling on and off through the night as it relights, those spikes add up more than people realise. Worth connecting a Victron SmartShunt if you haven't already — you'll see exactly where the Ah are disappearing rather than guessing from the BMS readout alone. @RustyTinker the Cairngorms in February is proper testing conditions, so don't be too hard on your setup!